Lexicon PCM80 AD-converter

 Finally I have found a replacement chip for CS5389 AD-converter in my PCM80 reverb. Lexicon worked from digital-input but analog inputs were silent. I once ordered CS5389-chip with eBay from China but that never arrived to Finland. Now I tried creative_dynamo from eBay and this was a success. I ordered direct replacement chip CS5390 from creative_dynamo UK. After I had checked the voltages and replaced the old 89-chip with 90 one everything worked as supposed. 

Now my PCM80 is back in business. Thanks creative_dynamo!

AS3046 DIP

Alfa's AS3046 DIP format chip is a direct replacement of old LM3046/CA3046 chip. Vbe is matched to +-1mV (LM3046 only to +-5mV). Another Moog Prodigy-clone is alive and with AS3046s the VCOs track  at least 8-octaves (as much as I can get from Arturia MicroFreaks CV-out).

AS3046's are here right the 14-pin DIPs.

This is the first version of my clone with old-style switches.

Joystick MIDI-controller

Today you can buy cheap Arduino UNO R3 microcontroller copy from Velleman for just 15€. With two analog joysticks, Arduino with code, two resistors, wire, connectors and a Hofnar cigarette box you can build a simple MIDI-controller to control four parameters with two hands (MIDI CC).

From UPIC to IanniX

You can find here a free PDF-download of a book of UPIC and Iannis Xenakis. UPIC was a computer/software system that enabled sounds to be generated from graphic structures. First working version of UPIC was created in 1977. Today there is an open source software IanniX that is based on Iannis Xenakis works.


Controlling audio from IanniX on Vimeo.

Arturia MicroFreak

Bought this little synth because it has a capacitive keyboard with CV, gate and pressure outputs. I consider it also cheap. It's actually very nice synth with 4-voice paraphonics, Mutable synthesis, modulation matrix, arpeggiator and sequencer. Here it's speaking with Speech-synthesis.

Sigma · MicroFreak Speaking

Voix Celeste is released

Voix Celeste is now released on Bandcamp and streaming services. Enjoy !

Voix Humana

Free taste of the 31.3.2020 MANU-album.

Voix Celeste

New solo-CD of MANU is coming out soon. It contains seven new compositions and one cover song. Voix Celeste (it's also an organ stop) is about silence and silent music. Silence is not just physical absence  of sound. It's a state of mind that music creates in the mind of the listener. Silence can stand for a definite kind of presence rather than absence. Cover picture is edited from Robert Fludd's (1574-1637) Sun de metaphysico macrosmi (Wellcome Library, Creative Commons license).

Cristal Baschet

What an magical instrument! Developed by Baschet brothers in the early 1950's. Here Peter Pringle plays his DIY version of this instrument and Ethervox Theremin. Basic sound is generated by dipping your fingers in water and rubbing them along a series of glass rods. These glass rods are connected to metal resonators that amplify the sound and create additive harmonics.



This other video shows two baschets (basic and bass versions) playing Erik Saties Gnossienne no1. Here you can see how wet those rods are. Interesting harmony between human, water, glass and metal.

Picturesque 21

Buchla inspired track. This started from exploration of Allen Stranges patch number 21.

Buchla 208 summa summarum

There was only one thing that didn't work as supposed in this build. Sequencer didn't operate correctly in the 3-stage mode (it worked as a 2 step sequence). 4 and 5 step stages worked ok. I changed the 4009 CMOS chip and after that everything worked as supposed!

Complex oscillators coarse slider useful range was only about the first half travel. I made the ModularSynthesis modification and added a 82k series resistor with R45 to reduce the gain of the coarse slider. That fixed this problem.

This build cost me about 2100€. Building was fun and the documentation in ElectricMusicStore, ModularSynthesis, Muff Wiggler and PortaBellabz helped a lot.

Buchla Bongo-box

I made a programming card where random-voltage changes sequencers last step (just one resistor). This way the 5-step sequencer sounds like a longer sequencer and is more interesting. I also switched both lopass-gates serial (signal routing). There is internally a 180-degrees phase-shift between them so when I in the middle of this clip rise the LPG2-level same as the LPG1-level there happens some random phase-cancellations (random voltage opens the LPG2). This sounds like a delay or velocity control. There is no external effect in this clip. Just the internal spring-reverb.

Metaprogramming the Electro Organism

I haven't enough banana cables so it is a good time to test the programming card. In Allen Stranges Buchla Easel manual there is a self-generating patch number 21.

I made a programming that based in this patch. Buchla is in both-mode (prog board + front panel). Random-voltage controls number-of-sequencer-steps, envelope-attack/decay and MO-modulation-mode.Sequencer controls MO-frequency and envelope generator controls lopass-gate1-level. Remaining connections are made with banana-cables. I adjusted the faders to my taste and here is the resulting sound-video.

Buchla calibration

Now Buchla is roughly in tune (1V/Oct) and calibrated. I changed R3 from 10k to 8.2k. This slighty decreases timbre folding and helped the calibration. The trickiest part was to balance the CO-timbre (motherboard TR4) and CO-waveshape (motherboard TR7) with Card9's TR5 (CO low freq track). The problem was to get pure sine-waveshape from bottom to top without any extra oscillations when applying the maximum timbre fold.

I calibrated CO and MO to 1V/Oct with trimmers TR8 and TR9 on motherboard. External CV must be connected to voltage input from keyboard banana-jack. Here is one sound example with long attack/decay times and with random AM-modulation of CO. Some external reverb is used.

Buchla is alive!

It's alive and creating sounds! Next I start to calibrate it. At the same time I check that everything is working. I love the outlook of this thing. It's like a synthesist's candy.

Colors of the sound!

Here in the middle are the calibration trimmers of motherboard.

Buchla front panel and motherboard

I found the front panel from The Beast. Everything fits great and this starts to look like a Buchla 208. Next I must finish the motherboard so that I can start to solder the banana jacks.

Buchla 208 status and the motherboard

All 12 cards are now ready. Next I will start the assembly of motherboard.

Following traces has been added to this 2.1 version of motherboard PCB:

- trace between Molex 4A and Molex 3B
- trace between Program Card pin7 and via
- trace to Card 8 pin 11 instead of pin 12 is corrected by connecting these two pins on motherboard
- trace between SW8 and SW10

To change the trigger threshold (Pulser, Sequencer, Envelope Generator) from 10V to 5V I will change R56 to 150k.

To enable my 208 to communicate with iProgram card I must do the following modifications:

- connectors B and 2 are connected together and grounded to Q (quiet/analog ground)
- pin C is connected to +5V
- pins D and 4 are connected together and grounded to N (noisy/digital ground)

Photo by Modular Synthesis

I have cloned this motherboard (and all the cards) with all the corrections. Maybe I will build some day another 208 from my own PCBs.

Buchla Card 12

I replaced the original Buchla Output-card with Boops Card 12 replacement. The reverb circuit is here better designed by driving the tank with current instead of voltage. I used the Davebr BOM from ModularSynthesis. With this card it is possible to bring out the MO and CO separately but I think I am not going to do this and I must jumper edge pins 7 and 9, let's see.

TOP

BOTTOM

Buchla Card 11

Card 11 is the LPG2. I changed C4 from 1uF to 47uF. This allows lower frequency MO (modulation oscillator) modulation of CO (complex oscillator). I also matched the vactrols VT1 and VT2.

Buchla Card 10

Card 10 contains the Preamplifier and LPG-1. Vactrols VT1 and VT2 needs to be matched. In electro-music forum I found an other matching method where vactrols are connected in series and LDRs form a voltage divider between +V and -V. In theory the more similar vactrols are the closer to 0V the output will be. Pulse LFO worked the best and I managed to match to pairs of vactrols for Card 10 and 11. I cross-checked this to the first matching method with Card 5 and these give similar results.

Vactrol matching.